Introduction to Descriptive Statistics

By Dr. Manas Kumar Pal

Statistical Methods

Statistical Methods

Descriptive Inferential
Statistics Statistics
Descriptive Statistics

1. Involves
• Collecting Data
• Presenting Data
• Characterizing Data

2. Purpose
• Describe Data
Inferential Statistical

1. Involves
• Estimation
• Testing of Hypothesis

2. Purpose
• Make decisions about
population characteristics
Descriptive Statistics

LEARNING OUTCOMES

• Compute and interpret measures of center and

spread of data

• Construct and analyze graphical displays to

summarize
Types of Descriptive Statistics

 Measures of central tendency or location or Averages

 Measures of variability or dispersion
 Graphs (measures of skewness & kurtosis)
Graphs

representations of data enabling the researcher to see

what the distribution of scores look like

graphic vs. numeric

 graphic may be better for visualization

 numeric are better for statistical/inferential purposes

Skewness (Lack of symmetry)
5

4 right
(positive) skew
3
X

0
0.0 0.2 0.4 0.6 0.8 1.0 1.2
5
D

4 left
(negative) skew
3
X

0
0.0 0.2 0.4 0.6 0.8 1.0 1.2
D
Kurtosis (Peakedness or Flatness)

Leptokurtic  Mesokurtic (Normal)  Platykurtic

Measures of Central Tendency

These are computed to give a “center” around which the

measurements in the data are distributed.

These are constant which concentrates to the central

part of the distribution

 Arithmetic mean or Mean

 Geometric Mean Mathematical
 Harmonic Mean average

 Median
Positional average
 Mode
Mathematical Average
Arithmetic mean

Sum of all measurements divided by the number of

measurements.
Use- For quantitative data

Geometric mean
Use- To find growth rate,
For construction of index number

Harmonic mean

Use- To find average speed

Positional Average

Median
the point in a distribution above and below which are 50% of the
scores
Use- Qualitative data

Mode

the score attained by more participants

Use- Business forecasting
Working Formulae
Measure Raw data Grouped data
s
Mean 

x

x
 fx  fx

x
n f N
 N 
Median (n  1)  (  C.F ) 
Q2  th observation, Q2  l   2
 f
h

2 



' n' is odd where l  lower lim it of median class
N  Total frequency
( n) (n  1)
Q2  th or th observation, C.F  Cumulative frequency just preceeding the
2 2 median class
' n' is even f  frequency of the median class
h  width of the median class

 ( f m  f1 ) 
Mode M 0  Highest ocuurence M 0  l   h
 m
2 f  f 1  f 2 

where l  lower lim it of mod al class

f m  frequency of the mod al class
f 1 and f 2  frequencie s just preceeding
and succeeding the mod al class
h  width of the mod al class
Working Formula

Measures Raw data

Weighted Mean 
X
 wx
, where w  weights
w
Combined Mean   
 n1 x1  n2 x2  ......  nk xk
X 
n1  n2  ......  nk
Points to remember
• When the Mean is greater than the Median the data distribution
is skewed to the Right.

• When the Median is greater than the Mean the data distribution
is skewed to the Left.

• When Mean and Median are very close to each other the data
distribution is approximately symmetric.

• Empirical relationship between mean, median & mode

Mean-Mode=3(Mean-Median)

• Relationship between Arithmetic mean(A.M), Geometric

mean(G.M) & Harmonic mean(H.M)
A.M≥G.M≥H.M
Measures of Dispersion
Measures of Dispersion...

Tells about the heterogeneity or homogeneity of

the distribution.
Measures of Dispersion…

 Range
 Quartile deviation
 Mean deviation
 Standard deviation
Range
Difference between two extreme values (H-L)
Use- Share Price , weather forecast
Quartile deviation
One half of the difference between the inter quartile
(Upper quartile – lower quartile)

Q.D= Q3  Q1 
1
2
Where  3N   N 
 (  C .F )   (  C.F ) 
Q3  l   4 h Q1  l   4 h


f 
  f 
 
   
Mean deviation

1 
Mean deviation taken from mean   x  x
n
For raw data
1 
  f xx
N
For grouped data
Standard deviation  
It is the positive square root of the deviation taken from
arithmetic mean

 

2
1   2
  1  x  
S tan dard deviation   
n 
 x  x 

 
 n
 x 2
  
 n  
   
For raw data
 

2
1   2
  1  fx  

N
 f 

x  x 

 
 N
 fx 2
 
 N  

   
For grouped data
Variance 
2

It is the square of the deviation taken from arithmetic

mean

 
 
2
1   2
  x 
Variance    x  x     x 2  
1
n   n n  
   
For raw data
 
 
2
  2
  fx 
  f  x  x     fx 2  
1 1
N   N N  
   
For grouped data
Coefficient of variation(C.V)

Used to find out efficiency of the distribution

 
(C.V ) 100   
 
x